Anat Levin<title> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> <meta name="keywords" content=""> <link href="/math/profile/design2.css" rel="stylesheet" type="text/css"> <script language="Javascript" src="/math/profile/footer.js" type="text/javascript"></script> <script language="Javascript" src="/math/profile/NavSet.js" type="text/javascript"></script> <script language="Javascript" src="/javascript/NavGen.js" type="text/javascript"></script> </head> <body> <div id="headleft"><img src="/math/profile/images/profile.gif"></div> <div id="headright"><img src="/math/profile/images/faculty.gif"></div> <div id="navigation"> <script language="JavaScript"> initHorNav(); horNav.displayNavHor("scientists"); </script> </div> <div id="content"> <body> <h2>Anat Levin</h2> <p> </p> <p><img align=left border=1 src="pictures/anatlevin.jpg" width="14%" "hspace=10" "vspace=10">  My research interests are in the fields of computer vision, computer graphics and machine learning. I am particularly interested in the emerging field of computational photography, which aims to take digital photography a step beyond the rigid imaging model inherited from traditional photography. As its name suggests, computational photography deals with computations, which facilitate image processing after the initial shot. These computations dictate certain aspects of the camera design, vastly expanding the amount and types of information that can be obtained from the picture during image processing. My colleagues and me have employed computational principles to help develop several innovative post-exposure applications: transparency, colorization, matting and segmentation. We have also applied these principles to the design of novel cameras, such as the coded aperture camera, in which a simple change to conventional lenses has made it possible to calculate depth from a single shot. Another simple innovation has helped produce a motion invariant camera, which can overcome motion blur distortions, bypassing the need to estimate scene motion and segment multiple scene motions. <p> </p> <p><strong>Recent Publications</strong></p> <ul> <li> [with P. Sand, T. S. Cho, F. Durand, W. T. Freeman] Motion-Invariant Photography. <em>SIGGRAPH, ACM Transactions on Graphics</em>, August 2008. <li> [with R. Fergus, F. Durand, W. T. Freeman] Image and Depth from a Conventional Camera with a Coded Aperture. <em>SIGGRAPH, ACM Transactions on Graphics</em>, August 2007. <li> [with W. T. Freeman, F. Durand] Understanding camera trade-offs through a Bayesian analysis of light field projections. <em>Proc. of the European Conference on Computer Vision (ECCV)</em>, Marseille, France, October 2008. <li> [with D. Lischinski and Y. Weiss] A Closed Form Solution to Natural Image Matting. <em>IEEE Trans. Pattern Analysis and Machine Intelligence</em>, February 2008. <li> [with D. Lischinski and Y. Weiss] Colorization using Optimization. <em>SIGGRAPH, ACM Transactions on Graphics</em>, August 2004. </ul> <p> </p> <p><a href="http://www.wisdom.weizmann.ac.il/~levina">Personal Web Page</a> </p> </div> <div id="footer"> <script language="javascript"> writeFooter(); </script></div> </body> </html>